Classification Using Improved Hybrid Wavelet Neural Networks

  • Authors:

  • Nhu Khue Vuong;Yi Zhi Zhao;Xiang Li

  • Affiliations:

  • School of Computer Engineering, Nanyang Technological University, Singapore 639798;Singapore Institute of Manufacturing Technology, Singapore 638075;Singapore Institute of Manufacturing Technology, Singapore 638075

  • Venue:
  • PRICAI '08 Proceedings of the 10th Pacific Rim International Conference on Artificial Intelligence: Trends in Artificial Intelligence
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this study, we propose a novel neural net-based classifier called improved Hybrid Wavelet Neural Networks (iHWNN). iHWNN makes good use of the characteristics of Wavelet Neural Networks (WNN) and Back Propagation Neural Networks (BPN), so that it inherits WNN's capability in learning efficiency and BPN's applicability in handling problems of large dimensions. To show the advantages of the developed algorithm, we compare its performance with those from existing classifier systems on several applications. Comparable results are achieved over several datasets from the UCI Machine Learning, with an average increase in accuracy from 91.69% for classification-based objective functions training to 94.17% using optimized iHWNN networks.